du33169/roberta-large-finetuned-GLUE-RTE

RTE

This model is a fine-tuned version of FacebookAI/roberta-large on the GLUE RTE dataset. It achieves the following results on the evaluation set:

• Loss: 0.8489
• Accuracy: 0.8231

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 64
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
No log	1.0	39	0.5271	0.6898
No log	2.0	78	0.7220	0.5576
No log	3.0	117	0.7798	0.4923
No log	4.0	156	0.7942	0.5525
No log	5.0	195	0.7762	0.7992
No log	6.0	234	0.8051	0.7044
No log	7.0	273	0.7204	0.8087
No log	8.0	312	0.8489	0.8231
No log	9.0	351	0.8779	0.8123
No log	10.0	390	0.9375	0.8159

Framework versions

• Transformers 4.43.3
• Pytorch 1.11.0+cu113
• Datasets 2.20.0
• Tokenizers 0.19.1